Product
Why Auno never acts without your yes
An AI that does real work has to earn trust one step at a time. Here’s how we designed approvals so Auno can move fast without ever surprising your team.
6 min read

Alexia Hart
Founder & CEO

Most AI tools are built to answer. Auno is built to act: it posts updates, drafts replies, files tasks and moves work forward across the tools your team already uses. That’s useful right up to the moment it does something nobody asked for.
So from the very first prototype, we set one rule: Auno never acts without your yes.
Answering is easy. Acting is a promise.
When an assistant only answers, a wrong answer costs you a few seconds. When it acts, a wrong action lands in a channel, an inbox or a customer’s thread. The stakes change, so the design has to change with them.
“The goal isn’t an AI that asks permission for everything. It’s one that knows when to.”
Alexia Hart, Founder & CEO
How approvals work in Auno
Every action Auno takes falls into one of three groups:
Read. Searching docs, summarising threads, pulling numbers. Auno does these freely, within what you can already see.
Draft. Replies, updates and tasks. Auno prepares them and waits for you.
Act. Posting, sending or changing anything other people will see. Always your call.
Trust you can turn up
Once you’ve approved the same kind of action a few times, Auno offers to handle it on its own next time, like posting the Monday update to #product. You stay in control: every automatic action is listed, and you can switch it back with one click.
It’s a little slower on day one. By week two, most teams barely notice the approvals, because Auno has learned which ones they actually want to see.
What’s next
We’re working on team-wide approval rules, so an admin can decide which actions need a second pair of eyes across the whole workspace. If you’d like early access, let us know through the contact page.
Share this post

Written by Alexia Hart
Founder & CEO at Auno. Alexia writes about building AI that teams can actually trust.

